Shipeasy

Plan limits & quotas

Per-plan knobs — poll interval, resource caps, retention, and which statistical features are available on each tier.

Production readyOn this page · 4 min readUpdated · July 4, 2026Works with · Free · Pro · Business

Plan levers are platform knobs we control centrally — they live in the plan config, not per project. Changing a plan value re-paces every project on that plan through the next KV rebuild, with no per-project migration.

Numbers here are illustrative

The tables below reflect current defaults but can move. Your dashboard shows your live limits and usage — check Project → Billing for the authoritative numbers and where you stand against them.

Read path

The biggest functional difference between tiers is how often the SDK refreshes its rules.

KnobFreeProBusinessEnterprise
Poll interval~300s~60s~30s~10s
Analysis frequencyDailyDailyDailyDaily
Propagation modelInfinite-TTL + explicit purgeInfinite-TTL + explicit purgeInfinite-TTL + explicit purgeInfinite-TTL + explicit purge

The poll interval is the worst-case lag between an edit going live at the edge and a given SDK instance noticing it on its next background poll. Evaluation itself is always local and instant — see Evaluation & caching. The Worker advertises the interval via X-Poll-Interval and the SDK auto-adjusts, so a plan change re-paces your fleet without a redeploy.

Resource caps

Flags, configs, and team members are unlimited on every tier — we cap usage (events), not capability. Unlimited means no plan-level cap (platform fair-use ceilings still apply for abuse protection).

ResourceFreeProBusinessEnterprise
Feature flags (gates)UnlimitedUnlimitedUnlimitedUnlimited
Dynamic configsUnlimitedUnlimitedUnlimitedUnlimited
Team membersUnlimitedUnlimitedUnlimitedUnlimited
Ops issuesUnlimitedUnlimitedUnlimitedUnlimited
Autopilot runs / moUnlimitedUnlimitedUnlimitedUnlimited
Running experiments31025Unlimited
Groups per experiment41020Unlimited
Universes31025Unlimited
Metrics25100500Unlimited
Event catalog entries502501,000Unlimited
SDK keys1050200Unlimited
i18n keys1,00010,00050,000Unlimited
i18n locale profiles2UnlimitedUnlimitedUnlimited

Included monthly usage & overage

Event usage is notify-only on paid plans — events are never dropped or rejected. When a project crosses 80% / 100% of its monthly credit, the usage cron notifies you (and ops, for paid projects). Flag evaluations are never metered. Free is hard-capped per day for abuse protection.

Allowance / rateFreeProBusinessEnterprise
Events / month1M (daily cap)20M, then $1 / M100M, then $0.75 / MCommitted
Flag evaluationsAnti-abuse capNever meteredNever meteredNever metered
Included ops seatsUnlimited3, then $7/seat5, then $12/seatCustom
AI-translated words5K50K250KCustom

Retention

DataFreeProBusinessEnterprise
Experiment results~30 days~90 days~180 days~365 days
Raw events (Analytics Engine)~90 days~90 days~90 days~90 days
Audit log~7 days~30 days~180 days~365 days

Feature availability

Statistics land at Pro; governance (SAML/SSO, RBAC, audit, holdouts) lands at Business — never behind an enterprise-only paywall.

FeatureFreeProBusinessEnterprise
SRM detection
MCP access
CUPED variance reduction
Sequential testing
Custom significance threshold
Change approvals
Holdout groups
Data / warehouse export
SAML / SSO
Custom RBAC roles
SCIM provisioning
Sequential testing & holdouts

Sequential testing lets you peek at results without inflating false positives — available from Pro. Holdouts live on the universe (a fixed segment excluded from every experiment in it), a Business feature. SRM detection (sample-ratio-mismatch guard) is on for everyone.

Changing plans

Upgrades take effect immediately and re-pace your fleet through KV — nothing to redeploy. Manage your plan in Project → Billing, where you'll also see live usage against every limit on this page.

Was this page helpful?
✎ Edit this page

On this page